Imagine a smart wheelchair that doesn't just move, but understands social cues to navigate crowds safely. It works with a team of flying drones and cameras that act like a digital lookout, guiding the user from door to door. The goal is to make these robots predictable and safe enough that people can trust them with their lives.
Current AI for autonomous vehicles is often a 'black box,' making it difficult to certify for safety in crowded areas. This prevents the widespread adoption of autonomous wheelchairs for people with reduced mobility.
A system comprising smart wheelchair hardware, aerial robots, and an orchestrator, along with a methodology to certify AI robustness.
If you are a medical device manufacturer dealing with the difficulty of certifying autonomous mobility aids — this project developed a certification methodology for AI-based vehicles that ensures safety and ethical compliance. This allows for the creation of smart wheelchairs that can operate autonomously in crowded areas.
If you are a facility management firm dealing with accessibility gaps in large public buildings — this project developed an orchestrator that links wall-mounted cameras and flying robots to guide wheelchairs. This creates a seamless, door-to-door experience for people with reduced mobility.
If you are an AI auditing firm dealing with the 'black box' problem of neural networks — this project developed runtime verification techniques and explainability methods. This enables the validation of object detection and speech-to-text models for safety-critical use.
Based on available project data, there is no specific pricing or cost information provided.
The project tested the system with 52 external participants and 16 trained operators, suggesting a move toward industrial validation, though full scale is not yet detailed.
Based on available project data, specific licensing terms are not mentioned, but the project focuses on releasing an engineering methodology for AI certification.
The system integrates four components: wheelchairs, aerial robots, wall-mounted cameras, and a central orchestrator.
The project period runs from 2022-10-01 to 2025-09-30, indicating it is currently in its final stages of validation.
The consortium is heavily industry-weighted with 7 industrial partners (54% ratio), including 5 SMEs. This suggests a strong focus on commercial viability and practical application across 7 different countries, balancing academic research from 4 universities and 2 research centers.
Contact SPINDOX LABS SRL in Italy for technical integration details.
Contact us to explore licensing the AI certification methodology for autonomous mobility.
